Digital Infuzion is supporting the development of innovative techniques and ground breaking approaches for the integration of biological and clinical data and the evolution of informatics methodology. Our pioneering methods foster easy interpretation and present the most complex and interdependent information in an intuitive fashion.  This facilitated interpretation can lead to better patient outcomes and improved understanding of research results.

Self-optimizing Software

Digital Infuzion’s proprietary data analysis software implements state-of-the-art machine learning, in which the software can learn from the data. Our patented approach to self-optimization enables these sophisticated mathematical modeling methods to be automatically fine-tuned without requiring user adjustments. GeneBrain® is a software tool that models large volumes of complex data for monitoring, classification of samples, repository querying, biomarker identification and the refinement and validation of hypotheses. GeneBrain® sits atop your data pipeline, integrating data from multiple platforms.


GeneBrain® provides visualization through a novel mathematical approach to pseudo-3D display of the samples in a space representative of the learning machine. These models are intuitively understandable by those without mathematical training and can be used for hypothesis formation and testing. Moreover, this novel graphical interface enables retrieval of data records by direct selection from the graphic.

Multiple Types of Data

GeneBrain® may be used on a single type of data, such as gene expression, or may be used to combine multiple types of data with automatic and appropriate integration of diverse data types such as protein expression, patient clinical tests, and SNPs. This enables the identification of small groups of biomarkers of mixed data types, even when single, simple biomarkers are not available.

Comparison Tool

GeneBrain® provides a repository comparison tool. It uses machine learning based on a pattern-matching algorithm to make comparisons. This provides a capability for analyzing multiple types of disparate biomedical data. Searches can be based upon particular sets of important data features, and relies directly on the data, not on the presence of annotations. This capability can be used to find similar samples or to identify similar groups of patients without preconceived ideas about what is likely to be similar.

Hypothesis Refinement and Validation

The comparison capability extends to hypothetical patterns. GeneBrain will retrieve the actual data records that are most similar to a hypothetical data pattern. This means that researchers can validate and refine their ideas in silico (on computer or via computer simulation) before expending laboratory funds.

Diverse Data

GeneBrain® analysis on integrated genotype (SNP), DIGE protein and immune factor data used 15 features drawn from four data streams to effectively characterize high (blue) or low (red) patient response to vaccine.

Related Clients